6. Dreaming and Behaviour

Orienting questions to think about

• what stages of sleep are associated with dreams?

• why do we dream?

• what is the relationship between dreaming and recent events?

  • why do we dream about recent events?

  • does dreaming help us retain memories

What is involved in primary consciousness?

• simple awareness, perception, emotion

• common to most mammals?

What is involved in secondary consciousness?

• self-awareness, abstract thought, “free-will” - unique to humans?

How do dreams relate to primary/secondary consciousness?

• dreams exhibit primary but not secondary consciousness

• for example, dreams often contain

  • visual components, sense of movement

  • strong emotions (e.g. anxiety, elation, anger)

• but we rarely notice or reflect on:

  • bizarre cognition (e.g., plot changes)

  • implausible/impossible elements (e.g. chimeric figures)

How has the scientific study of dreams evolved?

• for many years, relied solely on self-report

  • dream diaries, lab-based induction of dreams, awakenings at different parts of the night

• more recently, augmentation with polysomnography and neuroimaging add extra dimensions

  • but we still rely on participants’ dream reports to a great extent

• difficult to draw firm conclusions

  • dreaming remains deeply mysterious (yet fascinating)

  • very hard to know whether dreaming is epiphenomenal (just what happens when brain is in REM) or intrinsic component of sleep (has purpose)

What stages of sleep are associated with dreaming?

What did Dement and Kleitman (1957) find regarding whether REM sleep is associated with dream recall?

• woke participants at different times during sleep (REM and NREM)

  • asked for dream report

  • strong association with REM

• if people woken in REM sleep then they tended to recall a dream

• opposite is true for non-rem sleep

What did Picard-Deland et al (2023) find regarding what stages of sleep are associated with dreaming?

• sampled all stages of sleep

• but also sampled across early, mid and late sleep

• all stages showed dream recall, particularly REM

• but dream reports are longer for REM, particularly later in the night

• dreams rated for clarity the next morning, REM and N1 dreams were clearest

What is involved in Freud and Solm’s Psychodynamic theory of dreaming?

• Dreams fulfil unconscious wishes/desires

• REM sleep not a good model

• dream amnesia → repression

• function: preserve sleep in face of unconscious need for excitement/unknown

• dreaming is the day’s residue

What is involved in Hobson’s Activation-Input Modulation (AIM) theory of dreaming?

• three factors: activation, input and modulation

• REM sleep is a good model

• Dream amnesia due to REM sleep brain state dependence

• function: virtual reality model (protoconsciousness). Approach difficult problems in a safe place

• unrelated to the previous day

What is involved in Foulkes and Domhoff’s neurocognitive theory of dreaming?

• what occurs when brain is activated. Draws on memory schemas, general knowledge etc

• REM sleep not a good model

• dream amnesia → context effect, lack of external narrative

• no function → byproduct of the evolution of sleep and consciousness

• not related to previous experience apart from LTM

AIM model visualised, different sleep states

• conscious state determined by combination of the 3 variables: activation, input-out gating, modulation

• any brain state described as a point in this cube

• NREM → midpoint in the space, some brain activity, some modulation

• REM → just as much brain activity as during wake, little input/output gating, little modulation (protoconsciousness)

• lucid dreaming → higher than REM for input/output

What is involved in the idea of virtual reality/protoconsciousness?

general account

• general account: REM sleep offers opportunity to fine-tune “predictive model” of the world

  • update predictive model based on day’s sensory input

  • fits nicely with observation that newborns have large amounts of REM sleep (lots to update)

What is involved in the idea of virtual reality/protoconsciousness?

specific account: Threat Stimulation Theory (TST)

• consistent with activation of amygdala during REM sleep

• predicts that incidence of threatening events will be more common in dreams than everyday life

• a good opportunity for rehearsal of threat avoidance relevant to surroundings?

What did Abbas and Samson (2023 find regarding Threat Simulation Theory?

• analysed dream and wake reports during the COVID-19 pandemic

introspectively, it makes sense that dreams will have more threat than wake

• clear overrepresentation of negative events in dreams compared to waking life

• but do our dreams particularly involve avoidance practice?

What did Malcom-Smith et al (2012) find regarding whether there is a bias towards threat avoidance in dreams?

• categorised dream reports in terms of whether they were related to approach or avoid scenarios

• also looked at whether incidence changed for populations more or less likely to encounter life-threatening events (south-africans(more dangerous) vs welsh)

• found prevalence of approach dreams, plus no effect of environment

• more consistent with general protoconsciousness account than TST?

What have PET studies of REM sleep compared to wake and SWS found?

• activation of

  • visual areas (occipital cortex)

  • memory (medial temporal) and emotional (amygdala) systems

• deactivation of

  • inhibitory system (frontal lobe; cf, freud & lack of repression

• generally, activity in REM fits our understanding of dreams, but in many cases brain areas have more than one function - danger of selection bias

• can we “decode” visual areas of the brain during dreaming?

What is involved in dream reading?

Horikawa et al 2013

• only 3 ppts but collected lots of wakenings

• fell asleep in the scanner + eeg to determine stage of sleep

• about dreaming in N1

• woken up after short while and asked for dream report

• presented pictures from dream items to ppts in a waking fMRI to see what their brain signature is for that item - decoded to sleep volumes to see which of the objects are being dreamed about

• what extent do you beat chance for different brain areas

• experimenters were able to make good predictions about what the brain images were in sleep → the brain areas are processing those stimuli as if they were really there

• most activity for the item just before the moment of awakening - a snapshot

Dream reading summary

• clear neural link between visual imagery in dreams and actual visual perception in wake

• decoding dreams is an exciting prospect in terms of objective measure of dreams

• but fMRI is not very practical

  • more recent studies have focused on polysomnography

What did Walmsley et al 2010 find regarding the first two minutes of dreams?

• intensive training (several hours over 3 days) on immersive visuomotor task (Alpine Racer 2)

• ppts woken in early stages of sleep

• first 15 secs (wake or Stage I) dominated by directly related imagery

• beyond 75 secs (stage 1 or 2) dominated by indirectly related imagery

• cognitive replay very common in early sleep stages

• not captured by diary approach (forgotten)

What did Horowitz et al, (2020) find about whether early night dreams aid creativity?

• we know that waking from N1 can help with problem solving in the number reduction task but their study didn’t find a link with dreaming

• Targeted Dream Incubation (TDI) may help

  • Dormio - hand-worn device for tracking N1 onset

  • links to a laptop or phone to enable verbal dream reports and deliver instructions

  • can use the instructions to guide N1 dreaming (e.g., “remember to think of a tree”

• does this help creativity post-sleep?

• ppts repeat a cycle of waking after a period in N1, deliver verbal dream report, receive instructions “remember to think of a tree”

• subsequent to 45 mins of this cycle, ppts took part in creativity tests related to the tree theme: creative storytelling, alternative uses test, verb generation

• clear benefit of sleep + incubation, compared with sleep + no incubation and wake controls

• plus, participants who had the most incubation-related dream reports showed the most creativity

  • N1 dreaming associated with dream-related creativity

  • great example for experimental manipulation of dreams

What did Fosse et al (2003) suggest regarding why recent events feature in dreams?

• dreaming is not episodic replay

• 65% of spontaneous waking sleep reports contained recent life events (cf freuds day residue)
  but few (1-2%) were episodic in nature (detailed, contextualised)

• perhaps a more integrative or interactive role?

What did Picard-Deland show (2023) in their more detailed analysis of why recent events feature in dreams?

• lots of day-residue memories

• but some older episodic memories, plus semantic memories as well

• after controlling for sleep stage, recent memories tend to dominate early-night sleep

• distant memories feature more later in the night

• the integrative aspect may occur later in the night?

What did Blagrove et all (2011) find regarding the Dream-lag effect and sleep stages?

• when do events resurface in dreams? compared daily diary reports with dream reports to gain ratings of event incorporations in dreams (0-5 scale)

• examined dream similarity over a period of 2 weeks after events

• results showed evidence for lots of matches for dream reports the night after events (day residue)

• these matches diminish over the next few days, but then recur on days 5-7 (“dream lag effect”)

• a second study using awakenings replicated the dream lag effect for REM awakenings, but not s2

• hard to know why there is a u-shaped profile of dream incorporations for REM, possibly this is marking the latter stages of memory consolidation?

What did Picard-Deland & Nielson, 2021 find regarding TMR and the Dream-lag Effect?

• participants played immersive VR flying game containing sound sequences prior to a morning nap

• sound sequences played during nap in either REM or NREM

• researchers looked to see whether targeted memory reactivation (TMR) of the sounds impacted on later dream reports (outside the lab)

• dreaming of the lab environment was largely restricted to the nap period and unaffected by TMR

• later dreaming of the video game was influenced by TMR at different times

  • for TMR in REM, day residue effect. TMR makes the day’s event more salient?

  • for TMR in NREM, dream lag effect. TMR enhances likelihood of consolidation, reflected later in REM as a dream lag effect?

What did Wamsley & Stickgold (2018) find regarding whether dreaming offers a window on hippocampal replay?

• trained on the virtual maze task

• wired for polysomnographic recording and ask to report mentations at different points throughout sleep

• sleep ppts that directly incorporated the maze task into at least one dream report showed greater overnight improvement in both completion time and distance travelled, compared to those without direct task incorporations

What did Hudacheck & Wamsley (2023) find regarding whether task-related dreaming predicts subsequent memory?

• meta-analysis of the literature (16 studies, 45 effects)

• overall, there was a significant effect - task-related dreams were associated with better subsequent memory

• when broken down by REM vs NREM, only NREM benefit found

• task did not influence effect size

• overall, at least in NREM, dreaming seems to be a window on memory consolidation processes

What did Spano et al (2020) find regarding dreaming with hippocampal damage?

• dream reports after awakenings during NREM and REM for HC patients (N=4) and matched controls (N=10)

• dream frequency much lower for patients (despite no difference in “blank” dreams) and content of dreams was impoverished

• could be problem with incorporation of recent memories

• but also perhaps weaker reconstruction of intact memories (cf. scene construction theory)

Conclusions

• all stages of sleep involve some form of dreaming

• dreams reflect recent events and not so recent events in complex manner

  • swift descent into indirect incorporation - within a couple of minutes (wamsley)

  • recent memories dominate early-night dreams (Picard-Deland et al)

  • late night dreams feature recent and older memories, plus semantic knowledge

• What is dreaming for?

  • N1 dreams seem to promote creativity (Horowitz et al)

  • NREM dreams may be a window on memory consolidation (Wamsley)

  • REM as usual is the most difficult stage to classify

    • May be a form of protoconsciousness/sandbos for dealing with threat?

    • But REM also shows a dream lag effect, which could reflect completion of consolidation process (Blagrove)

  • As with sleep more generally, it’s unlikely that there is a single role for dreaming